From exabytes to zettabytes, the Cisco Global Cloud Index: Forecast and Methodology, 2010 – 2015 shows why data centre installers should be on cloud nine.

Networking leader Cisco has released its fi rst ever Global Cloud Index, which is an ongoing effort to forecast the growth of global data centre and cloudbased IP traffi c. The forecast includes trends associated with data centre virtualisation and cloud computing.

Released in November 2011, the Index estimates global cloud computing traffi c will grow twelvefold annually from 130 exabytes to a total of 1.6 zettabytes by 2015, a 66% compound annual growth rate. (One zettabyte is equal to a sextillion bytes or a trillion gigabytes. 1.6 zettabytes is approximately equivalent to 22 trillion hours of streaming music, 5 trillion hours of business web conferencing with a webcam or 1.6 trillion hours of online HD video streaming.)

In short, the Index predicts:

Annual global data centre IP traffic will reach 4.8 zettabytes by the end of 2015. At the same time, global data centre IP traffic will reach 402 exabytes per month.

Overall, data centre IP traffic will grow at a compound annual growth rate (CAGR) of 33% from 2010 to 2015.

The number of workloads per installed traditional server will increase from 1.4 in 2010 to 2.0 in 2015.

The number of workloads per installed cloud server will increase from 3.5 in 2010 to 7.8 in 2015.

By 2014, more than 50% of all workloads will be processed in the cloud.

Annual global cloud IP traffic will reach 1.6 zettabytes by the end of 2015. Global cloud IP traffic will reach 133 exabytes per month.

Global cloud IP traffic will increase twelvefold over the next 5 years. Overall, cloud IP traffic will grow at a CAGR of 66% from 2010 to 2015.

Global cloud IP traffic will account for 34% of total data centre traffic by 2015.

Evolution of data centre traffic

From 2000 to 2008, peer-to-peer file sharing dominated internet traffic. As a result, the majority of internet traffic did not touch a data centre, but was communicated directly between internet users.

Since 2008, most internet traffic has originated or terminated in a data centre. Data centre traffic will continue to dominate internet traffic for the foreseeable future, but the nature of data centre traffic will undergo a fundamental transformation brought about by cloud applications, services, and infrastructure.

By 2015, one-third of data centre traffic will be cloud traffic.

Figure 1 summarises the forecast for data centre IP traffi c growth from 2010 to 2015.

The internet may not reach the zettabyte era until 2015, but data centres are already there.

While the amount of traffic crossing the internet and IP WAN networks is projected to reach nearly 1 zettabyte per year in 2015, the amount of data centre traffic is already over 1 zettabyte per year, and by 2015 will quadruple to reach 4.8 zettabytes per year. This represents a 33% CAGR.

This data covers network data centres worldwide operated by service providers as well as private enterprises.

Most traffic stays within the data centre

Consumer and business traffic flowing through data centres can be broadly categorised into three main areas (see Figure 2):

Traffic that remains within the data centre.

Traffic that flows from data centre to data centre.

Traffic that flows from the data centre to end users through the internet or IP WAN.

In 2010, 77% of traffic remained within the data centre, and this will decline only slightly to 76% by 2015.

The ratio of traffic exiting the data centre to traffi c remaining within the data centre might be expected to increase over time, because video files are bandwidth-heavy and do not require database or processing traffic commensurate with their file size. However, the ongoing virtualisation of data centres offsets this trend.

Virtualisation of storage, for example, increases traffic within the data centre because virtualised storage is no longer local to a rack or server.

Transitioning workloads to cloud data centres

A workload can be defined as the amount of processing a server undertakes to execute an application and support a number of users interacting with the application.

The Global Cloud Index forecasts that for the transition of workloads from traditional data centres to cloud data centres the year 2014 is expected to be pivotal — when workloads processed in cloud data centres (51%) will exceed those processed in traditional data centres (49%) for the first time. Continuing that trend, we expect cloud-processed workloads to dominate at 57% by 2015 (see Figure 3).

Traditionally, one server carried one workload. However, with increasing server computing capacity and virtualisation, multiple workloads per physical server are common in cloud architectures. Cloud economics, including server cost, resiliency, scalability, and product lifespan, are promoting migration of workloads across servers, both inside the data centre and across data centres (even centres in different geographic areas). Often an end user application can be supported by several workloads distributed across servers. This can generate multiple streams of traffic within and between data centres, in addition to traffic to and from the end user.

Global cloud IP traffic growth

Data centre traffic on a global scale will grow at 33% CAGR (see Figure 4), but cloud data centre traffic will grow at a much faster rate of 66% CAGR (a twelvefold growth between 2010 and 2015 – see Figure 5).

By 2015, more than one-third of all data centre traffic will be based in the cloud. The two main causes of this growth are the rapid adoption of and migration to a cloud architecture and the ability of cloud data centres to handle significantly higher traffic loads. Cloud data centres support increased virtualisation, standardisation, automation and security. These factors lead to increased performance, as well as higher capacity and throughput.

Application readiness

As new models of service delivery and cloud-based business and consumer application consumption evolve, the fundamentals of network characteristics are essential. Fixed and mobile broadband penetration, download and upload speeds, and latency are indicators of readiness for delivery to and consumption from the cloud.

Furthermore, although speeds and latency are significant to all interested in assessing the quality of broadband services, they are not the only metrics that matter. Understanding basic broadband measures provides insight into which applications are most likely to benefit from faster broadband services for end consumers and business users. With business and consumer applications alike, advancements in video codecs, traffic optimisation technologies, and more, in addition to speeds and latencies, will lead to additional mechanisms to isolate speed bottlenecks at different points along the end-to-end paths and lead to other technical measures that will give a better understanding of how to deliver the best quality of experience.

The cloud readiness characteristics are as follows.

Broadband ubiquity: This indicator measures fi xed and mobile broadband penetration while considering population demographics to understand the pervasiveness and expected connectivity in various regions.

Download speed: With increased adoption of mobile and fi xed bandwidth-intensive applications, end user download speed is an important characteristic. This indicator will continue to be critical for the quality of service delivered to virtual machines, CRM and ERP cloud platforms for businesses, and video download and content retrieval cloud services for consumers.

Upload speed: With the increased adoption of virtual machines, tablets, and video conferencing in enterprises as well as by consumers on both fi xed and mobile networks, upload speeds are especially critical for delivery of content to the cloud. The importance of upload speeds will continue to increase over time, promoted by the dominance of cloud computing and data centre virtualisation, the need to transmit many millions of software updates and patches, the distribution of large fi les in virtual fi le systems, and the demand for consumer cloud game services and backup storage.

Network latency: Delays experienced with Voice over IP (VoIP), viewing and uploading videos, online banking on mobile broadband, or viewing hospital records in a healthcare setting, are due to high latencies.

Conclusion In terms of data centre and cloud traffic, we are firmly in the zettabyte era. Global data centre traffic will grow fourfold from 2010 to 2015 and reach 4.8 zettabytes annually by 2015. A subset of data centre traffi c is cloud traffi c, which will grow twelvefold over the forecast period and represent over one-third of all data centre traffic by 2015.

A key traffic driver as well as an indicator of the transition to cloud computing is increasing data centre virtualisation. The growing number of end user devices combined with consumer and business users’ preference or need to stay connected is creating new network requirements. The evolution of cloud services is driven in large part by users’ expectations to access applications and content anytime, from anywhere, over any network and with any device.

Cloud-based data centres can support more virtual machines and workloads per physical server than traditional data centres. By 2014, more than 50% of all workloads will be processed in the cloud. “Cloud and data centre traffic is exploding, driven by user demand to access volumes of content on the devices of their choice,” says Cisco vice president of product and solutions marketing Suraj Shetty.

“The result: greater data centre virtualisation and relevance of the network for cloud applications and the need to make sense of a dynamically evolving situation. “

The Cisco Global Cloud Index provides insight into this traffic growth and trends so that organisations can make strategic long-term decisions. We will continue to develop and release the Cisco Global Cloud Index on a regular and ongoing annual basis, contributing to ‘cloud readiness’ efforts worldwide.”

Overall, based on regional average download and upload speeds and latencies for business and consumer connections, all regions can support some level of cloud services. However, few regions’ average network characteristics are currently able to support the high-end advanced cloud apps.